© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 31-GEMMASCOPE-2-RES-65K
    7. 13557
    Prev
    Next
    INDEX
    Explanations

    hate speech and offensive content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_65k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     весьма
    0.56
     Державний
    0.54
     ممکن
    0.53
     ذریع
    0.52
     tertentu
    0.51
    வரின்
    0.50
     нередко
    0.50
     вскоре
    0.50
    
    0.50
     පිළිබඳ
    0.49
    POSITIVE LOGITS
     tiktok
    0.87
     shitty
    0.86
     covid
    0.83
     fucking
    0.82
     fucked
    0.78
     Tiktok
    0.74
     fuck
    0.72
     raping
    0.72
     scammers
    0.72
     shit
    0.71
    Activations Density 0.059%

    No Known Activations